Identifying lithofacies plays a central role in studying sandbody architecture and reservoir quality in fluvial reservoirs. Logging data is widely considered the most effective method for identifying subsurface lithofacies. Many machine learning methods have been developed to automatically identify lithofacies by analyzing the value or patterns of well logs. However, poor generalization of many classification models has resulted from a lack of exploration into the intrinsic relationship between lithofacies characteristics, data distribution characteristics, and classification model applicability. To address this problem, we conducted research on core description, logging curve sampling processing for layer data, and lithofacies identification using gaussian mixture model (GMM) and back-propagation neural network (BPNN) for a tight sandstone reservoir in the northern part of the Sulige gas field. We investigated the relationship between lithofacies characteristics, logging data distribution, and the performances of machine learning classification models. Based on this relationship, we developed a gaussian mixture model-backpropagation neural network hybrid classification model (GMM-BPNN). The results indicate that the logging curve sampling method reduced deviation caused by adjacent lithofacies influence, and made the lithofacies characteristics constrain the distribution characteristics of logging data, thus improving the application of GMM and BPNN. We observe that the distribution of logging data becomes more centralized as the thickness of certain lithofacies increases, thus improving the performance of the GMM applicable to the classification of centrally distributed data. Conversely, the distribution of logging data becomes more discrete as the thickness of certain lithofacies decreases, thus improving the performance of BPNN applicable to the classification of discretely distributed data. Furthermore, the GMM-BPNN (with an F1-score of 0.95) outperformed individual GMM (F1-score of 0.76) and BPNN (F1-score of 0.77). The hybrid classification model also shows better outcomes in the identification of complex lithofacies in other areas.

鄂尔多斯盆地苏里格东二区走滑断裂结构复杂,断裂控制下的流体差异富集规律及分布特征认识一直不清楚,严重制约了区内目的层有利目标优选及开发调整.针对上述存在的问题,本次研究在地震资料精细地质解释的基础上,开展断裂结构解析,明确断裂的分期、分段特征及其对流体分布的差异控制作用.研究结果表明:(1)研究区走滑断裂在剖面上存在单断类和组合类共2类5型剖面构造样式,平面上组合出线性、斜列式、叠置状、羽状和辫状5种平面结构样式;(2)走滑断裂经历了加里东期—早燕山期挤压、燕山中晚期走滑和喜山期拉张3期构造演化阶段,从深层到浅层断裂结构经历了单断型、花状、雁列式的演化,平面上根据断裂结构样式差异自西向东发育线性剪切段、派生剪切段和混合剪切段3段;(3)不同结构样式的断裂控制不同的流体差异富集模式,研究区共发育单断-封闭-稳产上升型、组合-开启-低产速降型、拉分-半开启-高产缓降型和剪切-半开启-低产缓降型4类流体差异富集模式,其中单断-封闭-稳产上升型和拉分-半开启-高产缓降型日产气量、累产气量均较高且生产稳定,是研究区天然气主要的富集区域.研究成果揭示了走滑断裂对致密砂岩储层流体差异富集的重要控制作用,为区内下一步开发调整指明方向.

针对鄂尔多斯盆地伊陕斜坡北部上古生界气水关系复杂、探评井成功率低等问题,通过对伊陕斜坡北部的地震资料进行精细解释,发现这一地区发育复杂的断裂体系.研究表明:断裂走向主要分为EW、NW、NE这3组,正、逆及走滑断层均有发育,由南向北断裂逐渐趋于密集.分析其演化可划分为3个期次:Ⅰ期形成于海西期挤压作用,Ⅱ期形成于燕山期挤压作用,Ⅲ期形成于喜山期走滑拉张作用.结合钻探成果,认为断裂活动可改善储层物性、提供运移通道,控制上古生界低幅度构造圈闭,而后期改造形成的通天断裂对原生气藏具有调整分配或破坏作用.

The deeply buried reservoirs of Wenchang Formation in the Lufeng Depression, Pearl River Mouth Basin, display strong heterogeneity, and the major controls for the development of high-quality reservoirs remain unclear. To address these issues, we conducted a series of experiment analyses, including petrographic microscope, scanning electron microscopy, and X-ray diffraction, and analyzed the impacts of sedimentation and diagenesis on the quality of deeply buried reservoirs. The results demonstrate that the sandbodies of subaqueous distributary channel and mouth bar deposited in lowstand systems tract (LST) and highstand systems tract (HST), as compared to the beach-bar and subaqueous fan sandstones deposited in transgressive systems tract (TST), have coarser grain size, higher quartz content, and lower muddy matrix content, which induced stronger anti-compaction capability, higher preservation of intergranular pore spaces, and thus better reservoir qualities. The reservoir types developed in subaqueous distributary channel and mouth bar are mainly types I, II, and III with medium-low porosity and low-ultra low permeability, while beach-bar and subaqueous fan mainly developed type III reservoir with low-porosity and ultra-low permeability. The reservoirs developed in E2w of the study area have undergone strong compaction, intense dissolution, but weak cementation. The subaqueous distributary channel and mouth bar reservoirs in LST are adjacent to Ew4 source rock in spatial distribution, resulting in strong organic acid dissolution, and developed numerous dissolved pores. The charging of hydrocarbons before deep burial further inhibited the later compaction and cementation and protects the preservation of residual primary intergranular pores and secondary dissolved pores. The combination of these factors leads to the development of the abnormally high porosity and high-quality reservoirs in LST. The results of this study reveal the genetic mechanism of deep, high-quality reservoirs in the rift basin and guide the selection of high-quality reservoirs in the later stage.

苏里格气田东二区二叠系石盒子组盒8段(以下简称盒8段)为典型的河流相致密砂岩储层,其强非均质性及复杂的储层结构导致该区面临"甜点"储层优选困难等关键技术瓶颈.为此,在分析盒8段储层岩相类型及组合特征、岩相约束下测井数据特征的基础上,建立了一种契合岩相及其组合特征、测井数据特征、人工智能算法原理的径向基—多层感知器神经网络联合模型,并开展了储层岩相的精确识别与表征研究.研究结果表明:①盒8段发育块状层理砾岩相、槽状交错层理粗砂岩相、板状交错层理粗砂岩相、板状交错层理中砂岩相、平行层理中砂岩相、交错层理细砂岩相、波状层理粉砂岩相、块状层理泥岩相8种岩相类型;②盒8上亚段曲流河相储层岩相密度偏小、岩相频率偏高、对应测井数据分布较分散,盒8下亚段辫状河相储层岩相密度偏大、岩相频率偏低、对应测井数据分布较集中;③建立的径向基—多层感知器神经网络联合模型识别准确率可达89.06％,相较于单一神经网络模型、交会图、主成分分析和决策树等方法识别准确率明显提高.结论认为,建立的径向基—多层感知器神经网络联合模型不仅克服了现有岩相识别方法准确率低且难以推广的缺陷,而且对实现河流相强非均质性致密砂岩储层高效开发具有重要意义.

苏77、召51区块柱塞排水采气措施自2012年至今累计应用213口,通过对柱塞井排水采气工艺应用效果的分析与评价,总结出了一套针对高含水气藏柱塞排采工艺适用条件及相应配套制度的实践运用方法,可有效指导井位优选与制度调整工作,本文主要针对无效井开展研究,落实水气比适用上限.

石英砂替代陶粒已成为降低储层改造成本有效手段,在致密油气藏开发中得到大幅推广应用,通过石英砂、陶粒分别在岩板、钢板上的对比试验及同井组、同层位的压裂试气对比分析证实,石英砂因抗压强度低破碎率高而致导流能力大幅下降,在低压、高含水气藏条件下,造成压裂液及地层水流动受阻,气井返排及生产能力下降,并加剧地层能量损耗.

南堡凹陷柳赞地区沙三段沉积期构造运动复杂,多幕构造—沉积演化导致层序地层格架及沉积充填样式时空差异大,区域上层序尺度的研究已无法满足研究区沉积、砂体精细表征的需求.本次研究聚焦沙三段中上部主力含油层系,通过陆相层序地层划分与对比技术,揭示了研究区三级层序内部准层序组级次的层序结构及沉积充填特征,明确了其对油气勘探开发的指导意义.研究结果表明:沙三1～3亚段整体可划分为2个三级层序SQ1和SQ2,其中,主力含油层III油组和II油组为SQ1层序构造宁静期湖退体系域的两个进积准层序组,垂向上随着沉积水体逐渐变浅,整体呈现扇三角洲"小平原大前缘"到平原、前缘均衡发育的沉积演化特征;I油组和0油组为SQ2层序构造活跃期湖侵体系域的两个退积准层序组,随着可容空间与沉积物供给之比逐渐增大,扇三角洲也逐渐由平原、前缘均衡发育演化成扇三角洲逐渐后撤,以滨浅湖、半深湖为主的沉积特征.横向上,边界断层的差异活动特征控制了相同沉积期沉积环境的横向变化.准层序组级次的沉积特征控制了研究区有利储盖组合的发育,构造宁静期湖退体系域中晚期(III油组和II油组)和构造活跃期湖侵体系域早期(I油组)发育的厚层储集砂体与上覆构造活跃期末期湖侵泥岩(0油组)形成了有利的储盖组合;准层序组的结构特征控制了准层序单元及其内部砂体、流体的分布规律,退积式准层序组下部砂体更发育,整体呈现下油上水的特征,准层序单元的顶部是有利储集砂体及油气的分布位置.通过本次研究,揭示了沙三段中上部主力含油层系0～III油组的层序结构及沉积充填模式,为后期砂体分布预测及岩性油藏勘探、开发提供了指导.

为了全面认识区块天然气成藏控制因素,气水分布规律,为产能建设目标评价和井位部署提供充足依据,有必要对区块内的断层分布特征进行分析研究,明确区块断层平面发育展布特征.通过区域构造分析、地震资料上断点的识别组合,再结合实钻井漏和投产井静动态特征等现象,最终确定工区中部发育一条大型断裂,命名为巴音—木肯断裂,该断裂为北部泊尔江—海子断裂的派生断层.其对工区原生气藏存在明显的破坏,也是浅部储层成藏的主要运移通道;下步井位部署应尽量避开断裂带位置,降低上钻风险;同时加强浅部地层的气测异常监测.

苏77、召51区块面积2007平方公里,位于伊盟隆起南部坡折带,具有典型的高岩屑、强非均质、低气饱、低压力、高含水致密砂岩气藏地质特征.投产气井产量低且普遍产水,水气比高达3.29m3/104m3.柱塞气举作为工区主力型排水采气工艺,自2012年至今已累计应用205口气井.探索科学合理的柱塞井生产制度,提高单井最终采收率成为柱塞气举工艺的关键.经过六年的探索总结,形成了一套科学的柱塞气井生产制度优化技术,为高产水气藏的规模开发提供了技术支持,实现了高产水柱塞气井在保证长期稳产的前提下最高效开发.

研究区的储层具有低孔隙度、低渗透率、低压力、低丰度和高含水饱和度等特点.束缚水饱和度与地层的孔隙结构、岩石性质及形成条件有关,所以对不同类型的储层,影响其束缚水饱和度的因素不同,求取方法也不同.针对研究区的实际情况,利用核磁共振资料、测井资料分别建立了3种计算束缚水饱和度模型:①孔隙度建立模型,该模型对低孔、低渗及孔隙结构复杂的储层适用性很差;②用多元线性回归求束缚水饱和度,该方法取得了较好的效果;③用孔隙结构指数建模,该模型与核磁分析吻合程度最高,效果最好.因此根据研究区储层特点,选择利用孔隙结构指数建立束缚水饱和度模型,大大提高模型准确度,为后期测井解释提供了可靠依据.

通过对研究区块岩心的扫描电镜、X衍射、黏土矿物分析、阳离子交换容量测量、铸体薄片及水分析等资料的研究发现,高束缚水饱和度是造成苏77山23段低电阻的根本原因,高矿化度地层水是直接原因,导电矿物影响微弱.由于低阻气层在电性曲线上与气水层、水层差异不明显,应用常规测井技术和解释方法很难有效识别,在解释过程中容易漏掉气层.提出用孔隙度重叠法、阵列感应测井法和束缚水饱和度法进行低阻气层的识别,与常规测井解释方法相比,解释精度大幅度提高,并将上述方法应用到苏里格气田中,取得了显著效果.

In the period of discovery and evaluation of Permian System and Carboniferous System igneous rock hydrocarbon reservoir in Santanghu basin,well log plays a positive part.A set of comprehensive evaluation technologies which are matched with igneous reservoir and oiliness have been formed.By application of well log technologies such as fast chromatogram,quantitative fluorescence,pyrogenation gas phase chromatogram and slight drilling, comprehensive evaluation of lithology,physical property,oil-bearing of igneous rock can be realized.Based on the evaluation of igneous rock,the applications of well log technologies are systematically summarized,which can provide technology reference for comprehensive evaluation of well log for the future.

Complicated lithology and abundant oil types of Carboniferous and Permian volcanic rocks cause the evaluation of oil and gas and water difficult in Santanghu basin. In order to increase the coincidence rate of logging lithologic profile and oil-gas interpretation and take advantage of oil-gas quantitative examination of logging technique, reasonable matching countermeasures of logging technique are proposed by summarizing the logging practices in recent yeas in this area. The method of combination of field logging and log curve response characteristics to identify volcanic lithology is formed, and the technical ideas using logging and well log to comprehensively evaluate reservoir are determined.

本文介绍了吐哈录井信息远程处理系统的开发应用情况。该系统采用CDMA及VSAT卫星通信等传输方式,实现了现场各种通信条件下录井数据到基地的传输;开发了"现场录井资料处理系统"、"基地资料处理系统"以及"网上录井信息服务系统"等一系列配套软件,建成了集"现场参数采集、远程数据传输、基地资料处理、网上信息服务"于一体的独具特色的吐哈录井信息服务体系。该系统于2005年开始推广应用,经过不断开发与完善,目前已在探井、滚动评价井、水平井测井施工中全面使用,网上信息服务系统也已经普遍应用于油田相关单位和部门。

For logging, using NMR technique to evaluate reservoir is an innovation to absorb and use technique theories of other industries for reference. Started with logging technique status and reservoir characteristics of Tuha oilfield, the basic principle of NMR technique and the correlation of logging parameters and lab conventional analytical physical property parameters are introduced. Combined case history, application effect of NMR technique in Tuha oilfield are analyzed by distinguishing types of reservoir, identifying reservoir fluid property, and establishing interpretation standards. By conclusion of technique advantages, extending application prospect of NMR is presented.

Both mud logging and wireline logging technologies have their own superiorities and limitations on hydrocarbon reservoir interpretation and evaluation, so, developing the comprehensive interpretation and evaluation study of mud logging and wireline logging technologies has important significance for improving interpretation coincidence rate. Taking viscous crude layers of TYK area from Tuha Basin as an object of study and combining samples and applied results analysis, the paper introduced three methods which used geochemical logging and wireline logging technology to evaluate and interpret oil-gas-water layers, they are accoustic logging interval transit time & pyrolysis hydrocarbon quantum interpretation and evaluation method (AC-P_g method), porosity & pyrolysis hydrocarbon quantum chart method (-P_g method) and apparent formation water resistivity & geochemical bright spot value chart method(R_ wa-M method). The applied results of the above methods are good, their interpretation coincidence rate is over 90 per cent. The research results of this topic have a certain guiding significance for interpretation and evaluation method study to different geologic characteristics, different oil properties and different types of hydrocarbon reservoirs.

In accordance with the characteristics of heavy oil reservoir in Tuha oilfield,this paper studies oil and water layer identification method and productivity prediction method on the basis of reservoir characteristic study combining with geochemical and logging data.Field application has attained the goal of oil and water layer identification and productivity prediction and has brought good economic and social benefit.

Started with practical logging in Tuha, the laboratory experiment of quantitative fluorescence technique before its being put into production and its achievement obtained are introduced. From identification of gas zone and slightly weak oil and gas indication, assessment of oil and gas layer with low resistivity, discrimination of reservoir fluid property and oil quality, establishment of interpretation criterion and chart and so on, application effect of quantitative fluorescence logging technique in Tuha oilfield was analyzed. Based on the summarization of experiences and analysis of influence factors, prospect of the quantitative fluorescence logging technique in Tuha oilfield was presented.

The target of log information platform construction is to realize data acquisition and data processing as well as interpretation and evaluation, and release the achievements on the network of information platform to integrate and share the data together. Not only planning and design of log information platform, but also system structure and achievement way as well as application effect are introduced.